The Care Profile tab on the Participant page has been completely rewritten. Previously it dumped raw JSON for most data types and had broken tables for medications and schedules. Now every category renders in a purpose-built format with collapsible sections, condition flags, and document downloads.
The final stretch of April did not produce a grand feature reveal. Instead, it showed Brett working at a different scale: clean up leftover interface noise, inspect why TokenWatch still was not telling a convincing cost story, confirm what models were actually in use, and then reduce the public-facing output of all that internal diagnosis to one brief security note.
A housekeeping session that tackled three related problems: an unreadable build output, recurring dark-theme colour bugs, and seven specific UI issues across the forum, messenger, and user tasks pages. The build went from 821 warning lines to a clean output, and a colour system was created to prevent the theme bugs from recurring.
The scheduling brain of RABS just had its biggest renovation since the project began. The metaphor names everyone learned -- workshed, loom, ribbon -- have done their job and are now retired. In their place sits core_engine: a born-clean Postgres schema with 19 tables, a dedicated instance API at /api/v1/engine, a fully-wired program creation wizard, and a working philosophy for how the past becomes truth. Twelve backend route files were rewritten to point at the new schema. Seventy-three real billing rates replaced the POC junk rows. The wizard now handles NF2F lines, participant photos, participant fees, click-to-edit time slots, and full draft management. None of this is theoretical anymore.
The ROC (Record of Care) extraction pipeline has been hardened against two data integrity problems that were causing duplicate records and orphaned document references. Running the extraction multiple times is now safe and idempotent.
Reggie and Henry have crossed the threshold from "they can chat" to "they can do things". The broker now lets them look up real Deputy rosters and return them in a human-readable shape, send SMS through an approval queue that actually executes when Brett approves, and pull a top-sheet support snapshot for any participant. The first end-to-end approval pipeline test ran today and worked: Reggie looked up Nicole Barton's Wednesday shift, queued an SMS to Brett, Brett texted Y AP0001, and the SMS landed on Brett's phone seconds later -- via Reggie's number, audited in tool_calls, the whole loop closed.
A comprehensive review of the Type 2 agent system (Reggie and Henry) was conducted against Factory's new droid-exec capabilities. The review identified what is working well, what could be improved, and a concrete upgrade path. No code changes were made -- this is a planning document for the next phase of agent development.
The User Tasks board is live. This is the complete guide to using it -- not just what the buttons do, but when to use each task type, how to set up your board for different workflows, and how to make the card widgets work for you.
April 25 did not look like a single product sprint. It looked like Brett turning the project back on itself and asking whether several recent stories were actually true. That meant reviewing the Type 2 agent runner architecture, re-reading recent blog posts for continuity, checking whether the profile review had gotten the risks right, and cleaning up a few smaller UI bugs that had surfaced after broader launches.
April 24 looked, at first glance, like a profile cleanup day. But the more the sessions read through the feature, the clearer it became that this was not mainly about tidying code. It was about understanding the gap between a surprisingly rich underlying care-profile data model and a frontend/API surface that still showed only part of it.